Time series forecasting using a hybrid ARIMA and neural network model

By G. Zhang. 2003

4,188 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Management Science and Operations Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Management Science and Operations Research

Citations in the corpus

Conditional Time Series Forecasting with Convolutional Neural Networks (2017, methodology): “Hybrid approaches using neural networks and econometric models have also been proposed, e.g. [29].”
A hybrid ARIMA and artificial neural networks model to forecast particulate matter in urban areas: The case of Temuco, Chile (2008, methodology): “Zhang (2003) tested a hybrid ARIMA and ANN model over three kinds of time series, and concluded that the linear and non-linear time series patterns in the combined model improved forecasting more than either of the models used independently.”
Forecasting monthly inflow with extreme seasonal variation using the hybrid SARIMA-ANN model (2017, methodology): “Therefore, linear models do not always provide satisfactory results for complex, real problems (Zhang 2003; Khashei and Bijari 2010).”
Traffic Flow Prediction via Spatial Temporal Graph Neural Network (2020, methodology): “• ARIMA: Auto-regressive Integrated Moving Average model [21, 42, 47] with Kalman filter is widely used in time series analysis.”
Different Approaches to Forecast Interval Time Series: A Comparison in Finance (2011, methodology): “A further study can be found in Maia et al. (2008), where ITS are forecasted by means of the centers and radii time series using ARIMA models, MLP (Zhang et al. 1998) and hybrid models ( Zhang 2003 ), which combine the ARIMA model and the MLP.”
A review of combined approaches for prediction of short-term wind speed and power (2014, background): “Hybridizing a linear model along with a nonlinear model to predict the hidden linear and nonlinear components embedded in the original wind speed or power signal, respectively with the aim of improving the prediction performance can be pointed out as a good example for the application of the hybrid ”
Modeling Long- and Short-Term Temporal Patterns with Deep Neural Networks (2017, methodology): “t the unknown time series in a look-ahead horizon – the large the horizon, the harder the problem. Efforts in this direction range from the early work using naive RNN models [7] and the hybrid models [16, 33, 34] combining the use of ARIMA [3] and 2 Multilayer Perceptron (MLP), to the recent combina”
Stock Market Forecasting Research Based on Neural Network and Pattern Matching (2010, methodology): “[9] Usually we take 5 days, 10 days, 20days or 60 days as technical analysis cycle.”
A Review of ARIMA vs. Machine Learning Approaches for Time Series Forecasting in Data Driven Networks (2023, positive): “In the classic work by Zhang et al. [4], a hybrid model combining ARIMA and neural networks is proposed, aiming to exploit the capabilities of each model in terms of linear and non-linear modeling, respectively.”
Natural language based financial forecasting: a survey (2017, methodology): “Some research adopts time series model like ARIMA [69], generalized autoregressive conditional heteroskedasticity (GARCH) [79], and combine it with machine learning techniques.”